Wiredframe Radar logo

Category intelligence

Social Media Briefing — March 16, 2026

357 current items analyzed and ranked.

Executive synthesis

Social Media Summary

A convergence of voices questioning deep learning's limits dominated the day. François Chollet argued the next breakthrough must come from below the model architecture level, revealing his startup Ndea is pursuing "symbolic learning". Gary Marcus penned a fiery open letter to Sam Altman claiming vindication, while Andrew Gordon Wilson critiqued next-word prediction as fundamentally opposed to creativity.

  • Ethan Mollick warned that AI bots have made social media comments completely unreadable, predicting humans will retreat to private spaces while the public web becomes a "bot wasteland"
  • Mollick also argued Meta and xAI have fallen behind frontier labs (OpenAI, Google, Anthropic), meaning recursive AI improvement will likely be proprietary
  • Swyx declared Qwen effectively over as an open-source powerhouse; Nathan Lambert explained skyrocketing costs are making open releases unsustainable
  • Meta reportedly shifting toward proprietary paid models, undermining Zuckerberg's open-source commitments
  • Demis Hassabis highlighted new AlphaFold applications, calling it "just the beginning of digital biology" — drawing massive engagement

Key Themes

Beyond Deep Learning / Need for New Paradigms · 10Open-Source Model Sustainability Crisis · 3AI Bot Pollution of Online Discourse · 10AI Lab Competition & Frontier Dynamics · 6AI Capabilities vs. Hype · 5Meta Open-Source Strategy Shift · 2Agentic AI & Workflow Automation · 2AI in Science & Biology · 3OpenClaw Ecosystem Expansion · 5Humanoid Robots & Robotics Competition · 14

Primary evidence

Top Ranked Signals

Social Twitter Mar 15

The next major breakthrough will branch out at a much lower level than deep learning model architect...

By @fchollet

82 score
AI Analysis

Chollet argues the next major AI breakthrough will require a fundamentally new approach below the level of model architecture, not just better architectures within the parametric learning paradigm.

The next major breakthrough will branch out at a much lower level than deep learning model architecture. It will be a new approach. A better model architecture can lead to incremental data efficiency & generalization gains, but it won't fix the fundamental issues of the parametric learning paradigm.
Beyond deep learningNew AI paradigmsLimits of parametric learning
Social Twitter Mar 15

I know I go on about this, but comments to all of my posts, both here and on LinkedIn, are no longer...

By @emollick

80 score
AI Analysis

Emollick's main post: comments on all his posts (Twitter and LinkedIn) are no longer worth reading due to AI bots, a dramatic change from just a few months ago. Describes bots as 'meaning-shaped attention vampires.'

I know I go on about this, but comments to all of my posts, both here and on LinkedIn, are no longer worth reading at all due to AI bots. That was not the case a few months ago. (Or rather, bad/crypto comments were obvious, but now it is only meaning-shaped attention vampires)
AI bot pollutionDead internet theorySocial media degradationFuture of online discourse
Social Twitter Mar 15

The failures of both Meta and xAI to maintain parity with the frontier labs, along with the fact tha...

By @emollick

78 score
AI Analysis

As discussed in Social yesterday, Emollick argues that Meta and xAI have failed to keep pace with frontier labs, and Chinese open-weights models lag months behind, meaning recursive AI self-improvement will likely come from Google, OpenAI, or Anthropic.

The failures of both Meta and xAI to maintain parity with the frontier labs, along with the fact that the Chinese open weights models continue to lag by months, means that recursive AI self-improvement, if it happens, will likely be by a model from Google, OpenAI and/or Anthropic
AI lab competitionRecursive self-improvementOpen-source AI gapFrontier model dynamics
Social Twitter Mar 15

i am actually still not over how Qwen as we knew it, one of the S tier Tigers, is over. https://t.co...

By @swyx

78 score
AI Analysis

Swyx expresses significant concern that Qwen (Alibaba's open model team), which he considered an S-tier 'Tiger' in open-source AI, is effectively over as we knew it.

i am actually still not over how Qwen as we knew it, one of the S tier Tigers, is over. t.co/Vj617GBDwv
open-source-modelsqwen-alibabaai-industry-dynamicschina-ai
Social Twitter Mar 15

Dear @sama, You owe me an apology. You have relentlessly, publicly and privately, attacked my inte...

By @GaryMarcus

75 score
AI Analysis

Gary Marcus writes an open letter to Sam Altman demanding an apology, arguing that Altman has now conceded what Marcus argued in his 2022 'Deep Learning is Hitting a Wall' paper: that current architectures aren't enough and something new is needed.

Dear @sama, You owe me an apology. You have relentlessly, publicly and privately, attacked my integrity and wisdom since my 2022 paper “Deep Learning is a Hitting a Wall”. But in your own way you have just come around to conceding *exactly* what I was arguing in that paper: that current architectures are not enough, and that we need something new, researchwise. beyond a scaling (a “megabreakthough” in your words below). That’s all I was trying to say. And I was right. And you should b
Scaling debateNew AI paradigmsBeyond deep learningAI industry discourse
Social Twitter Mar 15

if you can imagine it, you can build it

By @gdb

72 score
AI Analysis

Greg Brockman (OpenAI) posts 'if you can imagine it, you can build it' — massive engagement suggesting a major product/announcement context.

if you can imagine it, you can build it
AI capabilitiesOpenAI visionAI democratization
Social Twitter Mar 15

@swyx We're in the era when the cost of building LLMs is skyrocketing and the why for releasing them...

By @natolambert

72 score
AI Analysis

Continuing Lambert's Social analysis from two days ago, Nathan Lambert responds to Swyx's Qwen post, explaining that LLM training costs are skyrocketing while incentives for open release aren't changing. Mentions Llama may face similar issues but Nemotron is rising.

@swyx We're in the era when the cost of building LLMs is skyrocketing and the why for releasing them openly is static/not changing/weak. I'm working on making this more sustainable, but we should expect more seasons to this. First llama, then maybe qwen, but we have nemotron rising.
open-source-modelsai-economicsqwen-alibabanvidiaai-sustainability
Social Twitter Mar 15

Awesome non-coding Claude Code Loop ideas for business professionals. Grab any of these + immediat...

By @alliekmiller

72 score
AI Analysis

Allie K Miller shares creative non-coding use cases for Claude Code's /loop feature for business professionals, including email monitoring, meeting prep, competitor tracking, and proactive workflow automation.

Awesome non-coding Claude Code Loop ideas for business professionals. Grab any of these + immediately save time:
  • Check my email every 15min and ping me if something is related to Project Pluto and needs a decision made
  • Every 30min, prep me for my next meeting with attendee context, threads, mtg and crm notes, emails
  • Monitor a deal thread every 2h - summarize any new replies related to legal and suggest next moves
  • Research competitor announcements every 20min (better than an RSS feed b
agentic_aiclaude_codeai_productivityworkflow_automationai_use_cases
Social Twitter Mar 15

we used to have an internal goal, which felt totally impossible, of an AI that could write a coheren...

By @gdb

70 score
AI Analysis

Brockman reflects that OpenAI used to have an internal goal of AI writing a coherent 1000-line program, which felt totally impossible, and the tech has come a long way.

we used to have an internal goal, which felt totally impossible, of an AI that could write a coherent 1000-line program. the tech has come a long way!
AI coding capabilitiesAI historyOpenAI vision
Social Twitter Mar 15

> Meta shifts away from its open-source history toward a proprietary, paid model strategy Slippery ...

By @burkov

70 score
AI Analysis

Building on Reddit discussion of Meta's AI struggles two days ago, Burkov reports and critiques Meta shifting away from open-source toward proprietary/paid model strategy, calling Zuckerberg's open-source commitment conditional on financial benefit.

> Meta shifts away from its open-source history toward a proprietary, paid model strategy Slippery Zuckerberg "firmly believes in open-source" as long as it generates cash or quality hires. When cash runs out and hires aren't needed anymore, suddenly the belief in open-source is not as firm. Soft, rather.
Meta strategyOpen-source AIAI business models